Recently, I wrote about how I find the experience of browsing social media stressful, and, at the time, I expounded on Oliver Burkeman’s helpful insight—that the internet is a machine providing a limitless supply of things to worry about, in the face of which individual human finitude cannot help but be overwhelmed. That was particularly relevant to the specific contents I was seeing online then, but on the theme of online-induced anxiety, what I really always have on the back of my mind is Jia Tolentino’s essay The ‘I’ in the Internet, the first in her 2019 book Trick Mirror.
I first read it three years ago, and have since thought of it as the most definitive piece on the topic I’ve yet encountered. I’d forgotten some of the main points, so I decided to reread it. Below, I write down some notes and reproduce some of the most resonant highlights, mostly for my own benefit, to help affix the ideas in my mind. For you, dear imagined reader, of course I recommend reading the full text.
In the first section of the essay, Tolentino chronicles the evolution of the internet from its hopeful, innocent, promising Web 1.0 days, through to its descent into madness at the time of the essay’s writing in 2017–18, in the early years of the first Donald Trump presidency. Even now, in 2025, it feels like those years were when the internet’s potential for bringing about real-world chaos was fully made flesh. Things often feel like they’re still getting worse, but it was in those early Trump years when the floor gave way to the abyss. Tolentino describes an earlier inflection point in this arc of internet history:
As with the transition between Web 1.0 and Web 2.0, the curdling of the social internet happened slowly and then all at once. The tipping point, I’d guess, was around 2012. People were losing excitement about the internet, starting to articulate a set of new truisms. Facebook had become tedious, trivial, exhausting. Instagram seemed better, but would soon reveal its underlying function as a three-ring circus of happiness and popularity and success. Twitter, for all its discursive promise, was where everyone tweeted complaints at airlines and bitched about articles that had been commissioned to make people bitch. The dream of a better, truer self on the internet was slipping away. Where we had once been free to be ourselves online, we were now chained to ourselves online, and this made us self-conscious. Platforms that promised connection began inducing mass alienation. The freedom promised by the internet started to seem like something whose greatest potential lay in the realm of misuse.
At the end of the section, she lays out her diagnosis of the internet’s ills. This five-point framework is the source of this essay’s power:
In part out of a desire to preserve what’s worthwhile from the decay that surrounds it, I’ve been thinking about five intersecting problems: first, how the internet is built to distend our sense of identity; second, how it encourages us to overvalue our opinions; third, how it maximizes our sense of opposition; fourth, how it cheapens our understanding of solidarity; and, finally, how it destroys our sense of scale.
Regarding the first problem, she sketches the sociologist Erving Goffman’s seminal theory and metaphor of identity as playacting, the idea that life and social settings are a kind of theater in a very real sense. She gives several examples, one of which is that of the jobseeker, who in front of potential employers performs a well-rehearsed presentation of the identity of a highly qualified candidate, a conception of the self that is partly made up but which the jobseeker might actually come to fully believe in:
To communicate an identity requires some degree of self-delusion.
There is a wonderful definition of identity, one that always comes to my mind whenever I read particularly elaborate bios on social media profiles:
Identity, according to [the sociologist Erving] Goffman, is a series of claims and promises.
Currently, I often describe myself, among other things, as a recreational cyclist and somewhat of a cinephile, when asked what I am outside of work. I think about how I haven’t actually been on my bike for months now, and how few movies I’ve seen in that same timespan, and I think of false claims and broken promises.
People who maintain a public internet profile are building a self that can be viewed simultaneously by their mom, their boss, their potential future bosses, their eleven-year-old nephew, their past and future sex partners, their relatives who loathe their politics, as well as anyone who cares to look for any possible reason.
There is a particularly important idea here that touches also on cancel culture. L.M. Sacasas writes about the Age of the Database and how, given a sufficiently large amount of information available online about a person (true of celebrities and politicians and other high-profile people since perhaps the 20th century, and increasingly true of everyone else as well in the era of social media), it will always be possible to find contradictions and valid points of criticisms about that person.
(For the trivial reason that I just watched the new Superman film, I’m thinking of the filmmaker James Gunn, and how he was at one point cancelled, with disemployment consequences, because of some old provocative tweets that he has since disowned credibly. Now he’s an even more celebrated artist.)
Despite the huge amount of data transmission that the internet makes possible, and the tremendous transparency afforded by its ability to continuously expose private information to the public, the full complexity of a human person still cannot be faithfully reproduced online. The person that we get to know—or think we get to know—online is merely an illusion, because it’s always a distorted representation, a mediated performance.
Regarding the second problem, about how the internet encourages us to overvalue our opinions:
To try to write online... is to operate on... the assumption that speech has an impact, that it’s something like action; the assumption that it’s fine or helpful or even ideal to be constantly writing down what you think.
The third problem, I think, is obvious and also top-of-mind to most chronically-online people. The internet maximizes our sense of opposition through rage bait, the cloak of anonymity and the lack of the forces of presence and commitment in online spaces.
The fourth problem is particularly guilt-inducing. Solidarity online is cheap, Tolentino says. Hashtag campaigns are hollow:
A hashtag is specifically designed to remove a statement from context and to position it as part of an enormous singular thought.
Representation is not the same as the real thing:
It’s telling that the most mainstream gestures of solidarity are pure representation, like viral reposts or avatar photos with cause-related filters, and meanwhile the actual mechanisms through which political solidarity is enacted, like strikes and boycotts, still exist on the fringe.
Finally, regarding the internet’s inhuman sense of scale:
The internet was dramatically increasing our ability to know about things, while our ability to change things stayed the same, or possibly shrank right in front of us.
This is virtually the same point that Oliver Burkeman makes about human finitude in relation to caring; it’s just that Burkeman approaches it from the perspective of practical philosophy, while Tolentino approaches it from a more political-analytical point of view, if that is fair to say.
Near the end, she anticipates counter-arguments that worrying about new media technologies is something that thinkers and opinion-havers have done since ancient times. She starts with Socrates, who feared that the newfangled technology of writing might make people forgetful. Later on:
In the eighteenth century, men complained that newspapers would be intellectually and morally isolating, and that the rise of the novel would make it difficult for people—specifically women—to differentiate between fiction and fact.
She counters at the end with the observation that the internet is quite as absolute as it gets, that there is “nowhere further to go”:
Capitalism has no land left to cultivate but the self. Everything is being cannibalized—not just goods and labor, but personality and relationships and attention.
The conclusion is quite pessimistic, but there is advice nevertheless:
...we’ve got nothing except our small attempts to retain our humanity, to act on a model of actual selfhood, one that embraces culpability, inconsistency, and insignificance.
There is a list of suggestions that almost match the five fundamental problems one by one:
We’d have to care less about our identities, to be deeply skeptical of our own unbearable opinions, to be careful about when opposition serves us, to be properly ashamed when we can’t express solidarity without putting ourselves first.
Note that this is not the very end of the essay; the actual last sentence is something of a masterstroke, a bit of a bombshell, but I’ll leave that for those with copies of the essay to experience.
I’d also like to note, as a matter of precision, that in the essay, Tolentino is really talking about social media rather than the internet at large. There are still benign, useful places on the internet, about which these issues do not really apply. I don’t think I get anxiety issues while checking stocks of bicycle accessories on Decathlon’s online store, or Mass schedules on church websites—if they exist; unfortunately, most parishes just use Facebook instead, and in those cases then we’re back to our five problems.